In 2023, rare diseases accounted for 30% of product pipeline under development, about half of which comprising non-oncology rare diseases. Clinical development in rare diseases has specific challenges. A primary challenge arises from the small populations from which clinical trial participants can be recruited, and from often limited prior knowledge on a disease. The challenge of small populations may be exacerbated by a reluctance of some patients to enroll in the control arm of the trial. Limited knowledge of the disease’s natural history and heterogenous clinical manifestation for many rare diseases leads to difficulty in defining appropriate endpoints that are feasibly measured in clinical trials and to high uncertainty of assumptions that can be used to estimate a trial’s success rate. Effective clinical development strategy for rare diseases requires agility to adapt to accumulating learning.
Here, I provide a brief overview on the construction of endpoints, adaptive and seamless clinical trial designs, Bayesian approaches, and the use of natural history studies in the development of rare disease studies.
Endpoints construct
In some rare diseases, there is lack of appropriate endpoints that can be used in clinical trials. In cases of diseases with limited knowledge on their natural history, it is challenging to define and characterize the endpoint.
Another challenge comes from heterogeneity of the disease phenotype where it is not possible to have a single endpoint / outcome relevant for a heterogeneous group of patients. Composite endpoints could help address this challenge.
A number of statistical approaches can be used to construct and analyze composite endpoints. For some rare diseases with high unmet clinical need, surrogate endpoints can be used even for registrational trials.
If your clinical development strategy foresees a trial with a novel or surrogate endpoint, it is important that your plans include a generation of evidence to support such endpoint as fit-for-purpose, which can be based on external data as well as data generation in early phases of your therapeutic’s development. These plans need to be discussed with regulatory agencies as early as possible.
Adaptive and seamless designs
Adaptive designs with prospectively planned opportunities to modify study design elements can be especially helpful in rare disease trials when limited knowledge is available at the time of the study design.
Adaptive designs allow a possibility to stop for futility, stop for efficacy, or re-estimate the sample size at interim analysis.
Due to the heterogeneity of rare diseases, adaptive population enrichment designs are also used. These designs enable identifying which of the pre-defined subpopulations of patients are most likely to benefit from a treatment. At the interim analysis, a study can be stopped for futility, continue as planned with the overall population, or enrich the study population by enrolling new patients only in subgroups appearing to benefit from the new therapy.
Seamless phase 2/3 studies are also increasingly used in rare diseases. The operationally seamless approach allows the study to have the phase 2 and phase 3 stages under the same protocol, however, the population enrolled in the phase 2 stage of the trial will not be used for the analysis in the confirmatory stage. This approach can offer operational efficiency and save time, but it does not reduce the overall sample size. Inferentially seamless designs also have two stages under the same protocol, but the population enrolled in stage 2 will be included in the final analysis. Such designs allow for substantially reducing the sample size and shortening the program duration, however, they require a solid statistical methodology to meet the regulatory requirements of a registrational trial.
Adaptive trials can be designed within a frequentist or Bayesian framework. However, Bayesian trials may offer additional advantages for drug development in rare diseases.
Bayesian approaches
Bayesian designs may be more suitable for adaptive rare disease trials with an accelerated learning curve. Bayesian methods offer a natural way of seamlessly combining prior clinical information with accumulating data within a solid framework.
Bayesian borrowing means using data that are external to the trial vs. data generated within a trial. Usually, external data are only available for control arms, and most frequently, Bayesian borrowing is used to reduce the sample size and allows randomizing more patients to the active treatment. The Bayesian framework provides several methods that can be used for borrowing, such as robustified meta-analytic priors, power priors, and commensurate priors.
Another approach to minimize the number of patients assigned to the inferior treatment and to potentially reduce the sample size is response-adaptive randomization (RAR). In trials using the RAR approach, a chance of newly enrolled patients being assigned to treatment arms varies over the course of a trial and depends on accumulating data.
Single-arm trials are an extreme example where the effect of the investigational treatment is compared with the reference effect inferred from external data (external control). Because the absence of any concurrent control data and lack of randomization increases the risk of bias, the use of single-arm trials with external controls may only be supported in specific cases where the risks are deemed to be outweighed by the benefits.
Data from other clinical trials and real-world data (RWD) can be used for Bayesian borrowing and for external controls. RWD include data derived from EMR, disease registries, data generated from devices, or digital health technologies. Prospective and retrospective natural history studies (NHS) can be especially important to inform clinical development in rare diseases.
Natural history studies to inform clinical development
Natural history studies in rare progressive diseases can provide valuable information on the expected trajectory of the disease. NHS can be leveraged to inform clinical trial design in several ways. Informed by NHS, models describing the time course of the disease can be for internal decision-making. It can inform trial planning, inclusion/exclusion criteria, endpoints definition, and predict effect size and variability. Data from NHS can be used to build smarter analysis models, and patients from these databases may be easily incorporated into disease progression models as external controls.
NHS can be leveraged to construct external control arms for single-arm trials and for designs using Bayesian borrowing.
Each development program is unique and there is no single solution that can address all the needs. Exploration of several strategies, external data, and statistical trial simulations can help you enable data-driven decision-making to optimize your clinical development strategy.
Interested in learning more? Download our new ebook, Adaptive Trial Design, which outlines common adaptive trial designs, the benefits of adaptive trials, how to optimize your adaptive trial, and a ten-point framework to determine if your trial should be adaptive.
About Natalia Muehlemann
Natalia Muehlemann is Vice President, Clinical Development, at Cytel. Natalia Muehlemann, MD, MBA, has over 20 years of experience in general management, clinical development, and business development in the life sciences. Dr. Muehlemann joined Cytel in 2020, and prior to Cytel, served as Global Category Head, Acute Care - Oncology - Devices at Nestle Health Sciences. She acts as an Expert Jury member for the European Commission’s Innovation Council. Dr. Muehlemann holds an MD and an MBA (IMD) and professional certifications in statistics and data science.
Read more from Perspectives on Enquiry and Evidence:
Sorry no results please clear the filters and try again
Strategies to Overcome Limited Patient Population Challenges in Rare Disease Studies
Bayesian Strategies in Rare Diseases
Sample Size Re-Estimation for Rare Disease Clinical Trials
